获取结果失败，请稍后再试。
智能问答助手
结合文献知识库，保留通用大模型的能力，为您提供知识问答服务！
通用大模型
RAG知识库问答
Li, Rui ， Shi, Jiancheng ， Zhao, Tianjie ， Wang, Tianxing ， Lu, Shanlong
2020-01-01 null null null(卷), null(期), (null页)
Optical and thermal infrared remote sensing images highly integrate spatial heterogeneity information (land surface soil, vegetation and water). This paper evaluated the capacity of Landsat-8 and Moderate-resolution Imaging Spectroradiometer (MODIS) remote sensing indices and empirical relationship models for soil moisture estimations at different depths. The results show that (1) compared with other Landsat-8 indices, shortwave infrared based Surface Water Capacity Index (SWCI) has higher correlation with 10-50 cm depth soil moisture. The comparison based on MODIS daily indices confirms that SWCI can monitor 20 cm soil moisture with more stability; (2) The quadratic polynomial model based on Land Surface Temperature (LST) and SWCI possessed highest accuracy among all empirical models. The average coefficient of determination (R-2) increases to 0.257 from 0.150 based on LST-NDVI linear model and 0.176 based on LST-SWCI linear model. Soil moisture analysis at both 30 m and 1 km spatial scale suggest that optical remote sensing could indirectly reflect soil moisture variation with higher precise and more stability in root layer rather than top-most layer.